Androgens control the number of prostate cells by first, inhibiting cell death and inducing cell proliferation, and later by inhibiting further cell proliferation (shutoff). It has been postulated that defects in the shutoff mechanism are responsible for tumor development in aging human populations. Androgens also influence the proliferative behavior of prostate carcinoma cells; these cells display a variety of androgen sensitive and autonomous phenotypes. Hormone therapy has been used for triggering cell death and inhibiting cell proliferation; relapse occurs probably by selection for androgen-resistant phenotypes. Androgen- inhibition of cell proliferation is expressed by some prostate tumors; this phenomenon has not been fully explored despite its therapeutic relevance and potential relationship to the androgen-refractoriness of the mature prostate. The main goal of this research is to understand the mechanisms by which androgens inhibit the proliferation of prostate carcinoma cells; it is expected that this knowledge will provide new therapeutic tools. In addition, it may lead to the understanding of the proliferative quiescence of the adult prostate. Human prostate carcinoma LNCaP-FGC cells respond to androgens biphasically, in a time and dose- dependent manner. Variants LNCaP-TAC and LNCaP-LNO will be used to unravel these effects because the former expresses only the proliferative response to androgens whereas the latter expresses only the inhibitory effect. The working hypothesis states that the shutoff effect occurs because high androgen doses induce an intracellular repressor of the cell cycle. Specific Aims are: 1) To determine the time-course of the inhibitory response of LNCaP variants to androgens by means of flow cytometric analysis; this is needed to pursue the next aims: 2) To isolate putative mediators of the shutoff response by constructing subtracted cDNA libraries from shutoff+ and shutoff- mRNAs. 32P labeled cDNA inserts from differentially expressed clones will be used to probe mRNAs from all the variants by northern blots. Recombinants that are truly differentially expressed will be sequenced. Finally, 3) to assess the biological relevance of the isolated sequences. Mammalian expression vectors containing a constitutively expressed neo-resistance gene and the putative shutoff sequence driven by a metallothionein responsive promoter will be constructed and used to transform LNCaP-FGC and its variants; shutoff expression will be tested upon induction by Zn2+ or Cd2+. This research will shed light on faulty steps that lead up to the establishment of diverse hormone-sensitive and autonomous proliferative patterns in prostate cancer.